Published Paper Details:

Through this project, we aim to develop a Raspberry Pi-based smart home automation system that intelligently controls appliances by analyzing environmental conditions, occupancy status, and time of day using machine learning techniques. The system monitors indoor climate and evaluates ambient lighting conditions using environmental and light intensity data. For occupancy detection, the system replaces traditional IR sensors with a camera-based approach using Convolutional Neural Networks implemented via OpenCV, enabling more accurate and real-time detection of human presence within the monitored space. Additionally, time-based features including hour of the day and day/night classification are incorporated to enhance decision-making accuracy. A machine learning model is trained on historical data encompassing temperature, humidity, occupancy status and temporal features to predict optimal control actions for household appliances such as lights, fans, and air conditioners. The Raspberry Pi serves as the central controller, interfacing with sensors and the camera, and executing automation decisions in real time